The central goal of this project is to develop an innovative safe direct current (DC) stimulator to enable creation of neural prostheses that can excite, inhibit and modulate sensitivity of neural tissue. While effective in treating some neurological disorders, existing neural prostheses are limited because they can excite neurons but not efficiently inhibit them. This limitation stems from the need to use brief alternating current (AC) biphasic pulses maintain charge-balance across the metal-saline interface. DC current can both excite and inhibit neural activity; however, it is unsafe because it causes electrochemical reactions at the metal electrode-tissue interface. While cochlear and retinal prostheses successfully use AC pulses to encode sensory information by modulating firing rates of afferent fibers above their spontaneous activity, other neural prostheses fail to deliver effective treatment with excitation alone. For example, prostheses intended to assist micturition must excite sacral nerves to activate the detrusor muscle and simultaneously inhibit lumbar nerves to relax the urethral sphincter. For proper balance, inner ear vestibular afferent fibers require not only excitation to encode head motion toward the stimulated side of the head, but also inhibition to encode head motion away from it. Furthermore, several highly prevalent disorders characterized by high uncontrolled neural firing rates such as tinnitus, chronic pain, and epilepsy could be effectively treated by an implantable prosthesis capable of neural inhibition. One way to make DC safe involves directing DC flow of ions into the target tissue by switching mechanical valves of a saline-filled H-bridge in phase with AC square waves applied to electrodes immersed in the saline. This approach uses AC at the electrode-saline interfaces while maintaining DC ionic current through the tissue. Aim 1 is to build a functional safe DC stimulator (SDCS) prototype that can be controlled by a neural prosthesis to modulate graded cathodic and anodic stimulation currents in tissue. Aim 2 is to test the hypothesis that a neural prosthesis can use SDCS to control neural activity in vivo. In a very well-characterized chinchilla model of prosthetic stimulation of the vestibular nerve, we will compare performance of a neural prosthesis incorporating SCDS technology to that of a traditional vestibular prosthesis using biphasic pulse stimuli. Aim 3 is to test the hypothesis that SDCS stimulation is safe for chronic operation in the body. We will conduct constant chronic stimulation with three groups of implanted chinchillas to examine the durability and safety of SDCS. Groups receiving constant frequency pulsatile stimulation, constant anodic DC stimulation, and constant cathodic DC stimulation from the SDCS will be compared. We will assay responses to electrical modulation of the vestibular nerve in weekly sessions and ultimately by performing histological examinations.